Measuring robustness of brain networks in autism spectrum disorder with Ricci curvature

Anish K. Simhal, Kimberly L. H. Carpenter, Saad Nadeem, Joanne Kurtzberg, Allen W. Song, Allen Tannenbaum, Guillermo Sapiro, Géraldine Dawson

Abstract

Ollivier-Ricci curvature is a method for measuring the robustness of connections in a network. In this work, we use curvature to measure changes in robustness of brain networks in children with autism spectrum disorder (ASD). In an open label clinical trials, participants with ASD were administered a single infusion of autologous umbilical cord blood and, as part of their clinical outcome measures, were imaged with diffusion MRI before and after the infusion. By using Ricci curvature to measure changes in robustness, we quantified both local and global changes in the brain networks and their potential relationship with the infusion. Our results find changes in the curvature of the connections between regions associated with ASD that were not detected via traditional brain network analysis.

Introduction

Autism Spectrum Disorder (ASD) is fundamentally a disorder of brain connectivity, often characterized by atypical white matter development. While researchers use Diffusion Tensor Imaging (DTI) to map the brain's "wiring," standard metrics like fractional anisotropy often fail to capture the robustness of the network—how well it maintains function despite perturbations. This study explores the application of Ollivier-Ricci curvature (ORC), a concept from differential geometry, to quantify network stability in children with ASD undergoing umbilical cord blood infusion therapy.

Methods

The study analyzed data from a Phase 1 open-label clinical trial involving 19 children (aged 24–72 months) with confirmed ASD. Brain networks were modeled as weighted graphs using DTI data, where nodes represented 83 distinct brain regions. The core innovation lies in applying ORC to these graphs. In network science, positive curvature indicates a highly "triangulated" and robust neighborhood with redundant paths, while negative curvature suggests a more fragile, tree-like structure. Changes in curvature from baseline to six months post-infusion were correlated with clinical assessments, including the Vineland Adaptive Behavior Scales (VABS-SS) and the Expressive One-Word Picture Vocabulary Test (EOW).

Results

The ORC analysis identified significant improvements in network robustness that traditional tractography missed. These changes were specifically linked to clinical improvements in social and language skills.

Identified brain pathways (d, e, f) showing significant changes in Ricci curvature that were not detected by standard DTI metrics.

Key findings included:

  • Edge-Level Changes: Increased curvature in specific pathways, such as the connection between the right frontal pole and the right inferior temporal gyrus, correlated significantly with improved social and language scores.
  • Node-Level Changes: Scalar curvature (the average curvature of all connections to a node) increased in the left fusiform gyrus and the right pars orbitalis, regions central to the "social brain."
  • Sensitivity: ORC identified changes in the salience network (e.g., the connection between the dorsolateral prefrontal cortex and the insula) that traditional imaging techniques failed to capture.
A significant long-range connection where increased curvature post-infusion correlated with clinical improvement.

Discussion

The study treats Ricci curvature as a proxy for neuroplasticity. A more robust network (higher curvature) likely facilitates better signal processing, manifesting as improved behavior. The mathematical link between curvature and the number of redundant triangles at a node suggests that the therapy may encourage the development of supportive connections that stabilize neural communication. Unlike standard DTI measures, ORC showed a consistent relationship across multiple clinical measures, suggesting it is a more holistic indicator of patient improvement.

Conclusion

While limited by a small sample size and the lack of a placebo control, this study demonstrates that Ricci curvature is a sensitive metric for detecting subtle functional reconfigurations in the brain. Future work involving larger, double-blind trials will be necessary to validate ORC as a reliable biomarker for therapeutic efficacy in neurodevelopmental disorders.

Summary updated

Abstract

Ollivier-Ricci curvature is a method for measuring the robustness of connections in a network. In this work, we use curvature to measure changes in robustness of brain networks in children with autism spectrum disorder (ASD). In an open label clinical trials, participants with ASD were administered a single infusion of autologous umbilical cord blood and, as part of their clinical outcome measures, were imaged with diffusion MRI before and after the infusion. By using Ricci curvature to measure changes in robustness, we quantified both local and global changes in the brain networks and their potential relationship with the infusion. Our results find changes in the curvature of the connections between regions associated with ASD that were not detected via traditional brain network analysis.
BibTeX
@inproceedings{simhal2020measuring,
  author = {Anish K. Simhal and Kimberly L. H. Carpenter and Saad Nadeem and Joanne Kurtzberg and Allen W. Song and Allen Tannenbaum and Guillermo Sapiro and Géraldine Dawson},
  title = {{Measuring robustness of brain networks in autism spectrum disorder with Ricci curvature}},
  booktitle = {Scientific},
  year = {2020},
  doi = {10.1038/s41598-020-67474-9},
}
Figure